Shikonin was purchased from Sigma (St. Louis, MO, USA) and antibodies against MMP-2 and MMP-9 were purchased from Calbiochem (Boston, MA, USA). Antibodies against PCNA and p65 were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). MMP-9 ELISA kits were purchased from R&D Systems (Abingdon, UK).

Human breast cancer cell lines, MCF-7 and MDA-MB-231, were used in the present study. Cells were grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS).

The invasion of tumor cells was assessed in Transwell chambers. A Transwell membrane (8-μm pore size, 6.5-mm diameter; Corning Costar Corporation) was used. A Transwell membrane coated with Matrigel (100 μg/ml, 100 μl/well) was used for the invasion assay. Cells were seeded onto the upper wells in the presence of different concentrations of shikonin. The bottom chambers of the Transwell were filled with conditioned medium. After incubation for 24 h, cells were fixed, stained and counted under a light microscope.

The MMP activities were assayed as previously described (19). Briefly, 5×10⁵ cells in a 6-well plate were seeded and maintained in serum-free medium for 16–24 h. The conditioned medium was separated on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel containing 1 mg/ml gelatin. The gel was washed with buffer I [Tris-HCl (pH 7.5) and 2.5% Triton X-100)], incubated overnight in buffer II [150 mM NaCl, 5 mM CaCl₂, 50 mM Tris-HCl (pH 7.6)] at 37°C and stained with Coomassie blue. The clear bands indicate where MMPs degraded gelatin.

Migration was assessed by a wound healing assay. Cells were seeded at 2×10⁴ MCF-7 and MDA-MB-231 cells/well. After scraping the cell monolayer with a sterile micropipette tip, the wells were washed with serum-free medium several times with different concentrations of shikonin. The first image of each scratch was acquired at time zero. At 24 h, each scratch was examined and captured at the same location and the healed area was measured.

Cells were washed twice with 1X PBS, and cell extracts were prepared using RIPA buffer (1X PBS, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS containing an additional 100 ml of 10 mg/ml PMSF, and 1 Complete Mini Protease Inhibitor tablet). Lysate proteins were resolved by SDS-PAGE and transferred using nitrocellulose membranes (Whatman Protran BA83; 0.2 μm) The membranes were incubated with TBS buffer containing 0.1% Tween-20 and 5% skim milk, and then exposed to the desired primary antibody. After treatment with the proper secondary antibody, the immunoreactive bands were visualized using standard ECL (Pierce, Rockford, IL, USA) method.

In the reverse transcription-PCR (RT-PCR) analysis, total RNA was extracted from the treated cells. For reverse transcription reaction, cDNA was synthesized from 1 μg of total RNA using the High Capacity cDNA synthesis kit (Applied Biosystems). The PCR primers used were as follows: MMP-9 sense, 5′-TTTGACAGCGACAAGA AGTGG-3′ and MMP-9 antisense, 5′-TCCCATCCTTGAAC AAATACA-3′; MMP-2 sense, 5′-CATTCCGCTTCCAGGG CACAT-3′ and MMP-2 antisense, 5′-GCTCCTGAATGCCCT TGAGTCA-3′; GAPDH sense, 5′-GCCATCGTCACCAAC TGGGAC-3′ and GAPDH antisense, 5′-CGATTTCCCGCT CGGCCGTGG-3′. PCR products were analyzed by agarose gel electrophoresis and visualized by treatment with ethidium bromide.

The oligonucleotides were labeled with [γ-³²P] ATP and incubated with nuclear extracts for 30 min by using the Gel Shift Assay kit (Promega Corp., Madison, WI, USA). The DNA-protein complexes were resolved on non-denaturing and non-reducing 6% acrylamide gels. The probes used for EMSA were as follows. Probe for AP-1, 5′-CGC TTG ATG ACT CAG CCG GAA-3′.

The results are presented as mean ± SE, and statistical comparisons between groups were carried out using one-way ANOVA followed by the Student’s t-test. P≤0.05 was considered to indicate a statistically significant result.

To investigate the effects of shikonin on human breast cancer cell migration and invasion, we treated MDA-MB-231 cells with different concentrations of shikonin (1–5 μM) and performed wound healing and Matrigel-based Transwell invasion assays. As shown in Fig. 1, shikonin significantly inhibited MDA-MB-231 cancer cell migration by ~80% and invasion by ~60%. Moreover, shikonin induced a concentration-dependent inhibition of PMA-induced migration and invasion of MCF-7 cells. MCF-7 cells treated with PMA presented a 1.5-fold increase in migration (Fig. 1A and B). However, treatment with 5 μM shikonin inhibited PMA-induced migration by ~70% in the MCF-7 cancer cell line (Fig. 1A and B). We next examined whether shikonin inhibits PMA-induced invasion in MCF-7 cells. PMA induced a 2-fold increase in the invasion of MCF-7 cells, while shikonin inhibited this invasion by ~60% at a 5 μM dose (Fig. 1C and D). These data imply that shikonin is an effective inhibitor of breast cancer cell migration and invasion.

MMP-9 and MMP-2 are important ECM-degrading enzymes that are reported to be involved in cancer cell invasion and metastasis (20). On the basis of these observations and the foregoing data, we examined the effect of shikonin on the expression of both MMP-2 and MMP-9 in MDA-MB-231 cells; we also investigated PMA-induced MMP-9 expression in MCF-7 cells. Following treatment with shikonin, MMP-9 expression in the MDA-MB-231 cells gradually decreased over time at both the mRNA and protein levels (Fig. 2A and B). Shikonin also significantly inhibited PMA-induced expression of both MMP-9 mRNA and protein in the MCF-7 cells. However, neither PMA nor shikonin treatment affected MMP-2 mRNA or protein expression (Fig. 2A and B). We next examined MMP-2 and MMP-9 secretion and activity using gelatin zymography and enzyme-linked immunosorbent assays (ELISAs), respectively. MMP-9 secretion measured in the MDA-MB-231 cell-conditioned medium was significantly reduced by shikonin (Fig. 2C and D). Similarly, shikonin inhibited PMA-induced MMP-9 activity in the MCF-7 cells. These results indicate that shikonin is involved in the regulation of proteolytic activity as well as expression of MMP-9.

To explain the molecular mechanism underlying the inhibitory effects of shikonin on MMP-9 expression, we investigated the possible regulation of MMP-9 at the transcriptional level. The MMP-9 promoter contains two AP-1 binding sites and a nuclear factor (NF)-κB binding site, and it has been shown that NF-κB and AP-1 play critical roles in regulating basal and cytokine-induced MMP-9 expression in many cancer cell lines (19). Therefore, we examined whether shikonin inhibits MMP-9 expression through an effect on NF-κB or AP-1 binding sites in the MMP-9 promoter. Human breast cancer cells, MDA-MB-231 and MCF-7, were transiently transfected with the MMP-9 promoter construct. As shown in Fig. 3A, MMP-9 promoter activity was decreased in a dose-dependent manner following shikonin treatment in MDA-MB-231 cells. We also, confirmed that the MMP-9 promoter was activated up to ~10-fold in the PMA-treated MCF-7 cells. Shikonin showed a dose-dependent inhibition of PMA-induced MMP-9 promoter activity in the MCF-7 cells (Fig. 3D). Next, we examined whether shikonin affects the transcriptional activity of NF-κB and AP-1, critical modulators of MMP-9 activity. MDA-MB-231 cells were transiently transfected with AP-1 report vectors and then treatment of shikonin was carried out at different doses. As shown in Fig. 3C, AP-1 promoter activity was significantly reduced in the range of 1–5 mM of shikonin. In addition, MCF-7 cells were transiently transfected with reporter vectors containing tandem repeats of the AP-1 binding site to confirm the specificity of the shikonin-mediated inhibitory effect on AP-1. Treatment with PMA led to an ~7-fold increase in AP-1 promoter activity; notably, this stimulatory effect of PMA was significantly reduced by treatment with shikonin (Fig. 3F). However, NF-κB promoter activity exhibited significant change in the MDA-MB-231 and MCF-7 cells (Fig. 3B and E). These results suggest that shikonin inhibits MMP-9 activity through suppression of AP-1-mediated MMP-9 transcription in breast cancer cells.

Next, we examined whether shikonin inhibits the transcriptional binding activity of AP-1 to its DNA motifs in MDA-MB-231 cells using electrophoretic gel mobility shift assay (EMSA). MDA-MB-231 cells were treated with different dose of shikonin for 24 h and nuclear extracts were prepared. As shown in Fig. 4A, shikonin significantly decreased AP-1 DNA binding activity. Subsequently, we examined whether the subunit of AP-1 transcription factor is regulated by shikonin. Expression levels of c-Fos and c-Jun, subunits of AP-1, were significantly reduced by shikonin treatment (Fig. 4C). Similar results were observed in MCF-7 cells. MCF-7 cells were pretreated with different concentrations of shikonin with 100 nM PMA. Then, nuclear extracts were prepared and analyzed for AP-1 DNA binding activities. As shown in Fig. 4B, shikonin significantly decreased PMA-induced AP-1 DNA binding ability. To determine which subunit of the AP-1 transcription factor is regulated by shikonin, we examined the expression levels of c-Fos and c-Jun following shikonin treatment. Our data showed that shikonin significantly reduced PMA-induced c-Jun and c-Fos expression but had little effect on the expression of p65 in the western blot assays (Fig. 4D). These results suggest that shikonin inhibits AP-1 transcription activity by suppressing its subunit expression.